Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    phrases related to correcting, fixing, and resolving issues or mistakes

    oai_token-act-pair · gpt-3.5-turbo

    actions related to correcting or fixing problems

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    rogens
    -0.70
    ograms
    -0.70
    ãģ®æ
    -0.66
    ãģ®
    -0.66
    ertodd
    -0.65
    yi
    -0.65
    yes
    -0.65
    ï¸
    -0.64
    amara
    -0.63
    leans
    -0.62
    POSITIVE LOGITS
    balance
    0.94
     tensions
    0.82
     deficiencies
    0.82
     grievances
    0.81
     misconceptions
    0.80
    ulate
    0.77
     debts
    0.76
     stalled
    0.76
     longstanding
    0.75
     overcrowd
    0.75
    Act Density 0.296%

    No Known Activations